A method for optimizing processing parameters of fiber reinforced polymer composites considering the influence of tool wear in the whole milling cycle on machinability

Abstract

The application discloses a kind of fiber reinforced polymer composite processing parameter optimization method considering the influence of milling whole cycle tool wear on machinability, which includes: first, the multiple regression fitting model of cutting force, processing quality, delamination damage about cutting progress is established, then the constraint condition is constructed with the requirements of machine tool, tool and workpiece, then according to the fitting model and constraint condition, the cutting parameter optimization model with the goal of maximizing cutting efficiency and tool life is constructed, finally the improved ball rolling optimization algorithm is used to calculate the optimal parameters of fiber reinforced polymer composite (FRP) milling process, and the results are output in the form of pareto chart.This method fully considers the influence of tool wear progress on machinability such as milling force, surface roughness, milling delamination damage during FRP milling cycle, on the other hand, the algorithm adopts the strategy of combining local search and global search, with the characteristics of fast convergence speed and high solution accuracy.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of mechanical cutting processing, and relates to a machining parameter optimization method, in particular to a machining parameter optimization method for fiber reinforced polymer composites considering the influence of tool wear in the whole milling cycle on machinability. BACKGROUND

[0002] Fiber reinforced polymer composites (FRP) are ideal materials for high-performance products such as aerospace, aircraft, automobiles and sports equipment due to their high strength-to-weight ratio, high stiffness-to-weight ratio and strong designability. FRP is a difficult-to-machine material, which has low interlaminar shear strength, poor thermal conductivity, and is prone to delamination damage and other defects during processing. In addition, the fiber has high hardness and strong abrasiveness, and the tool wear is obvious as the processing progresses, which further affects the workpiece processing quality, such as increasing the surface roughness and aggravating the delamination damage. Since the processing quality and tool wear are both affected by the cutting process parameters, it is an important problem to select appropriate cutting parameters to achieve the comprehensive optimization of processing efficiency, quality and tool life (the number of qualified workpieces or material removal per tool).

[0003] However, most of the existing machining parameter optimization researches start from the performance of new tools, and the influence of tool wear progress on milling force, surface roughness, milling delamination damage and other machinability is not considered, so it is difficult to achieve optimization for the whole FRP processing process or considering tool life. In addition, the current algorithms for cutting parameter optimization all use meta-heuristic algorithms, and the parameter settings of these algorithms have a great influence on the solution results. Moreover, the existing technology rarely uses multi-objective algorithms based on non-dominated solution sorting to solve the cutting parameter optimization problem. SUMMARY

[0004] Therefore, in the processing of fiber reinforced polymer composites, the optimization is performed by the Scarabaeidae algorithm, and the optimal solution is selected by non-dominated solution sorting, so as to achieve the purpose of ensuring processing efficiency and processing quality, which has important significance for the processing and manufacturing of fiber reinforced polymer composites.

[0005] In order to overcome the above-mentioned defects existing in the prior art, the present application aims to provide a machining parameter optimization method for fiber reinforced polymer composites (FRP) considering the influence of tool wear in the whole milling cycle on machinability, which takes cutting processing efficiency and tool life as optimization objectives, establishes a cutting parameter optimization model according to the performance parameters of the machining tool used and the processing quality requirements of the workpiece in engineering, and further considers the influence law of tool wear progress in the whole processing cycle of fiber reinforced polymer composites on milling force, processing surface roughness, milling delamination damage and other machinability during the model establishment process.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0007] A method for optimizing processing parameters of fiber reinforced polymer composites considering the influence of tool wear on machinability in the whole milling cycle, comprising the following steps:

[0008] (1) Constructing a multiple regression fitting model of cutting force, processing quality, and delamination damage about cutting progress according to experimental data of cutting processing;

[0009] (2) Determining the constraint conditions of the fitting model according to the performance of the numerical control machine tool and the surface processing quality requirements of the processed part;

[0010] (3) Constructing a multi-objective cutting parameter optimization model in combination with the fitting model in step (1) and the constraint conditions in step (2);

[0011] (4) Solving the cutting parameter optimization model by applying an improved dung beetle optimization algorithm.

[0012] Further, the step (1) specifically comprises the following sub-steps:

[0013] (1.1) Establishing a fitting model of the surface roughness in the milling process according to the axial cutting depth a p , the radial cutting depth a r , the feed per tooth f z , the spindle speed n of the numerical control machine tool, and considering the influence of tool wear on the surface roughness as L progresses in the whole milling cycle:

[0014]

[0015] wherein C1, a1, a2, a3, a4, a5 are constants, and R a is the surface roughness;

[0016] (1.2) Establishing a fitting model of the delamination damage factor in the milling process according to the axial cutting depth a p , the radial cutting depth a r , the feed per tooth f z , the spindle speed n of the numerical control machine tool, and considering the influence of tool wear on the delamination damage factor as L progresses in the whole milling cycle:

[0017]

[0018] wherein C2, b1, b2, b3, b4, b5 are constants, and D is the delamination damage factor;

[0019] (1.3) Establishing a fitting model of the surface roughness in the milling process according to the axial cutting depth a p , the radial cutting depth a r , the feed per tooth f z, spindle speed n and considering the influence of tool wear on milling force with L advancing in the whole milling cycle, a fitting model of milling force in the milling process is established:

[0020]

[0021] In the formula, C3, d1, d2, d3, d4, d5 are constants, and F is the milling force.

[0022] Further, the step (2) specifically comprises the following sub-steps:

[0023] (2.1) determining the spindle speed constraint of the numerical control machine tool:

[0024] n min ≤n≤n max

[0025] In the formula, n min and n max are the minimum and maximum allowable spindle speeds of the machine tool or engineering;

[0026] (2.2) determining the feed per tooth constraint of the numerical control machine tool:

[0027] f zmin ≤f z ≤f zmax

[0028] In the formula, f zmin , f zmax respectively represent the minimum and maximum allowable feed per tooth of the machine tool or engineering;

[0029] (2.3) determining the axial depth of cut constraint of the numerical control machine tool:

[0030] a pmin ≤a p ≤a pmax

[0031] In the formula, a pmin and a pmax respectively represent the minimum and maximum allowable axial depth of cut of the machine tool or engineering;

[0032] (2.4) determining the radial depth of cut constraint of the numerical control machine tool:

[0033] a rmin ≤a r ≤a rmax

[0034] In the formula, a rmin and a rmax respectively represent the minimum and maximum allowable radial depth of cut of the machine tool or engineering;

[0035] (2.5) determining the constraint of surface roughness of the workpiece machined by the numerical control machine tool:

[0036] R a ≤R amax

[0037] wherein R amax represents the upper limit of the required surface roughness of the workpiece;

[0038] (2.6) determining the constraint of the delamination damage factor of FRP milling:

[0039] D≤D max

[0040] wherein D max represents the upper limit of the delamination damage factor of FRP milling;

[0041] (2.7) determining the constraint of milling force:

[0042] F≤F max

[0043] wherein F max represents the maximum cutting force allowed by the feed mechanism of the machine tool.

[0044] Further, the step (3) specifically comprises constructing a parameter optimization model as follows:

[0045]

[0046] wherein MRR = z x a p x a r x f z x n, z is the number of tool teeth; V = a p x a r x L, is the tool life objective function characterized by the material removal amount.

[0047] Further, the step (4) specifically comprises the following sub-steps:

[0048] (4.1) setting the basic parameters of the algorithm, including the number of populations N p , the number of solutions in the final output pareto solution set N, the proportion of the four-part population in all populations, the current iteration number t and the maximum iteration number T, and simultaneously applying a chaotic sequence to generate an initial population Q t , t = 0, as follows:

[0049] x n+1 = r x x n x (1 - x n ), x n ∈ [-1, 1], r = 4

[0050] xid = Lb + (Ub - Lb) x y id , d = 1, 2, 3, 4, 5

[0051] where x n is the position of the nth individual, Ub and Lb are the upper and lower bounds of the dth dimension of the problem solution search space, y id is the dth dimension of the ith individual position, x id is the dth dimension coordinate value of the ith individual position in the problem solution search space;

[0052] (4.2) Calculate the corresponding fitness of each individual position in the population Q0, check and mark whether each individual in the population is dominated by other individuals;

[0053] (4.3) Apply the traditional MBO algorithm to update the individual positions in the population, and divide the individuals in the population Q t into the following four types according to their order in the population: ball pushing MBO, brood MBO, small MBO, and thief MBO, and perform different strategies for updating the four types of MBOs;

[0054] (4.3.1) The position updating method of the ball pushing MBO is:

[0055] x i (t+1) = x i (t) + a x k x i (t-1) + b x Ax Ax = |x i (t) - x w |

[0056] where t represents the current iteration number; x i (t) represents the position of the ith MBO at the tth iteration; a is randomly selected as 1 or -1; k takes 0.1; b takes 0.3; Ax is used to simulate the change of light intensity; x w represents the global worst position;

[0057] After encountering an obstacle, the position updating method is:

[0058] x i (t+1) = x i (t) + tanθ x |x i (t) - x i (t-1)|

[0059] where θ ∈ [0, π], when θ is 0, π / 2 or π, the MBO does not update the position;

[0060] (4.3.2) The position updating method of the brood MBO is:

[0061] Lb *= max(X * × (1 - R), Lb)

[0062] Ub * = min(X * × (1 + R), Ub)

[0063] B i (t + 1) = X * + b1× (B i (t) - Lb * ) + b2× (B i (t) - Ub * )

[0064] where X * represents the current local best position, Lb * and Ub * represent the lower and upper bounds of the oviposition area, R = 1 - t / T, T represents the maximum number of iterations, Lb and Ub represent the lower and upper bounds of the optimization problem, B i (t) is the position information of the ith brood-care beetle at the tth iteration, b1 and b2 represent two independent random vectors of size 1 × D, and D is the dimension of the optimization problem;

[0065] The position updating method of the small scarab beetles is as follows:

[0066] Lb b = max(X b × (1 - R), Lb)

[0067] Ub b = min(X b × (1 + R), Ub)

[0068] x i (t + 1) = x i (t) + C1× (x i (t) - Lb b ) + C2× (x i (t) - Ub b )

[0069] where X b represents the global best position, Lb b and Ub b represent the lower and upper bounds of the optimal foraging area, x i (t) represents the position information of the ith small scarab beetle at the tth iteration, C1 represents a random number subject to a normal distribution, and C2 represents a random vector belonging to (0, 1);

[0070] The position updating method of the thief scarab beetles is as follows:

[0071] x i (t+1)=X b +S×g×(x i (t)-X * |+|x i (t)-X b |)

[0072] Wherein x i (t) represents the position information of the i th stealing beetle at the t th iteration, g is a random vector of 1 x D obeying normal distribution, and S takes 0.5;

[0073] (4.4) judge whether the solution set corresponding to the position of each individual in the new population exceeds the limit of the cutting parameter and take corresponding measures, then calculate the fitness of the position of all individuals in the new population and judge whether each individual is dominated by other individuals, record the dominated individuals and the dominated individuals of each individual;

[0074] (4.5) merge the population Q t And Q t-1 And the individual position of all individuals in the existing pareto solution set, compare all non-dominated solutions recorded in steps (4.2) and (4.4), find all non-dominated solutions after merging the population, then sort all non-dominated solutions according to the crowding distance, take the first N non-dominated solutions and put them into the pareto solution set, and these solutions are regarded as the optimal pareto solution set;

[0075] (4.6) check whether the maximum iteration number T is reached, if not, repeat steps (4.3) to (4.6) until the maximum iteration number is reached, at this time, the pareto solution set is the optimal pareto solution set.

[0076] Compared with the prior art, the beneficial effects of the present application are:

[0077] The present application comprehensively considers the influence of tool wear progress on machinability such as milling force, machining surface roughness, milling delamination damage in the whole machining cycle of fiber reinforced polymer composites, can improve the current problem of relying on experience to select process parameters, and realize high production efficiency and long tool life under the requirement of workpiece quality. The present application adopts chaotic initialization population, introduces non-dominated sorting and crowding distance sorting strategy to the traditional beetle optimization algorithm, effectively improves the distribution and convergence of the algorithm, and makes the obtained non-dominated solution set more uniform and accurate. This algorithm has the characteristics of beetle optimization algorithm itself on one hand, that is, it adopts the strategy of combining local search and global search, which can avoid falling into local optimum, on the other hand, makes the machining parameters can be selected according to actual needs. BRIEF DESCRIPTION OF DRAWINGS

[0078] Figure 1 is the overall flow chart of the fiber reinforced polymer composite processing parameter optimization method of the present application;

[0079] Figure 2 is the algorithm optimization process flow chart of the fiber reinforced polymer composite processing parameter optimization method of the present application. DETAILED DESCRIPTION

[0080] In order to make the purpose, technical scheme and advantages of the present application more clear and explicit, the present application is further described in detail below by examples, and in combination with the drawings. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application.

[0081] As shown in Figure 1 , a fiber reinforced polymer composite processing parameter optimization method considering the influence of tool wear in the whole milling cycle on machinability includes the following steps:

[0082] (1) Construct a multiple regression fitting model of cutting force, machining quality, and layered damage about cutting progress according to experimental data in cutting processing:

[0083] (1.1) Workpiece machining quality: Since the workpiece surface roughness is an important indicator to measure the machining quality of the part surface, the surface roughness is selected to describe the machining quality. Under the conditions of fixed tool geometry parameters and machine tool processing system, it is generally believed that there is an exponential relationship between the workpiece surface roughness of cutting processing and the cutting parameters. In addition, since the machining surface roughness of the workpiece is not constant in the milling process, it is affected by the tool wear in the whole milling cycle with the progress of L, which should also be an influencing factor in the calculation method of the machining surface roughness. Finally, the present application selects the profile arithmetic average deviation R a as the quality index of workpiece machining, so the machining surface roughness fitting model is:

[0084]

[0085] where R a is the machining surface roughness, C1, a1, a2, a3, a4, a5 are constants, a p is the axial depth of cut, a r is the radial depth of cut, f z is the feed per tooth, n is the spindle speed, and L is the feed distance.

[0086] (1.2) Milling Delamination Damage: FRP milling delamination damage refers to the debonding phenomenon that occurs between fiber layups. This phenomenon severely reduces the fatigue resistance of the structure and is the main reason for FRP scrapping. This invention uses a delamination factor to represent the degree of FRP delamination defects. In addition, this invention also considers the influence of tool wear as L progresses throughout the milling cycle on the milling delamination damage factor. Therefore, the calculation method for the FRP milling delamination damage factor is as follows:

[0087]

[0088] In the formula, D is the milling delamination damage factor, and C2, b1, b2, b3, b4, and b5 are constants.

[0089] (1.3) Milling force: Given the tool geometry and machine tool performance parameters, the milling parameters and milling force are related exponentially. Therefore, an empirical formula for milling force is established using an exponential form. In addition, the influence of tool wear as L progresses throughout the milling cycle on the milling force cannot be ignored. Therefore, the calculation method for milling force is as follows:

[0090]

[0091] In the formula, F is the milling force, and C3, d1, d2, d3, d4, and d5 are constants.

[0092] (2) Determine the constraints of the fitting model based on the performance of the CNC machine tool and the surface machining quality requirements of the machined parts.

[0093] (2.1) Spindle speed constraint: The cutting parameters of a CNC machine tool are closely related to the spindle. Therefore, when selecting cutting parameters, some constraints on the spindle speed must be met, namely:

[0094] n min ≤n≤n max

[0095] In the formula n min and n max These are the minimum and maximum permissible speeds for machine tool spindles or in engineering applications.

[0096] (2.2) Feed per tooth constraint: During CNC machining, the feed per tooth must be between the minimum and maximum allowable feed per tooth of the machine tool, i.e.:

[0097] f zmin ≤f z ≤f zmax

[0098] In the formula f zmin f zmax These represent the minimum and maximum allowable feed per tooth in a machine tool or engineering project, respectively.

[0099] (2.3) Axial depth of cut constraint: the axial depth of cut during milling process should be within the range of parameters allowed by the machine tool, i.e.,

[0100] a pmin ≤a p ≤a pmax

[0101] where a pmin and a pmax represent the minimum and maximum axial depth of cut allowed by the machine tool or engineering.

[0102] (2.4) Axial depth of cut constraint: the axial depth of cut during milling process should be within the range of parameters allowed by the machine tool, i.e.,

[0103] a rmin ≤a r ≤a rmax

[0104] where a rmin and a rmax represent the minimum and maximum axial depth of cut allowed by the machine tool or engineering.

[0105] (2.5) Surface roughness constraint: the surface roughness of the part should meet the minimum surface roughness requirement of the part quality, i.e.,

[0106] R a ≤R amax

[0107] where R amax is the upper limit of surface roughness.

[0108] (2.6) Milling delamination damage factor constraint: the milling delamination damage factor of FRP should be less than the upper limit of FRP milling delamination factor, i.e.,

[0109] D≤D max

[0110] where D max represents the upper limit of FRP milling delamination damage factor.

[0111] (2.7) Milling force constraint: during the milling process of the machine tool, when the tool turns one tooth, the workpiece and the tool will generate a feed resistance in the direction of the feed motion, and the feed resistance cannot exceed the maximum cutting force allowed by the machine tool feed mechanism, i.e.,

[0112] F≤F max

[0113] F max represents the maximum cutting force allowed by the machine tool feed mechanism.

[0114] (3) Combined with the fitting model in step (1) and the constraint condition in step (2), a multi-objective cutting parameter optimization model is constructed, specifically:

[0115]

[0116]

[0117] Wherein MRR = z x a p x a r x f z x n, is the machine tool cutting efficiency objective function, z is the number of tool teeth, and the machine tool cutting efficiency is represented by the volume of material removed per unit time in the present application; V = a p x a r x L, is the tool life objective function represented by the material removal amount.

[0118] (4) The process of solving the cutting parameter optimization model by applying the improved scarab optimization algorithm is shown in Figure 2 , and the details are as follows:

[0119] (4.1) Set the basic parameters of the algorithm, including the number of populations N p , the number of solutions in the final output pareto solution set N, the proportion of the four-part population to the total population, the current iteration number t, and the maximum iteration number T, etc. At the same time, the initial population Q t is generated by applying a chaotic sequence, t = 0, and the specific implementation is as follows:

[0120] (4.1.1) The improved scarab optimization algorithm adopts real number coding mode, and there are five optimization variables, spindle speed n, feed per tooth f z , axial depth of cut a p , radial depth of cut a r , and feed distance L, so for an initial population of N p individuals, each individual position can be represented in the form of a five-dimensional vector, that is:

[0121] Y i = (y i1 , y i2 , y i3 , y i4 , y i5 ), i ∈ [1, N P ], y id ∈ [-1, 1], d = 1, 2, 3, 4, 5

[0122] (4.1.2) First, randomly generate an individual position in the form of a five-dimensional vector according to step (4.1.1), and bring each dimension of the individual position into the following Logistic chaotic mapping equation for iteration. After N p -1 iterations, the remaining N p -1 individual positions are obtained:

[0123] x n+1 = r x x n x (1 - x n ), x n ∈ [-1, 1], r = 4

[0124] (4.1.3) Map the N p five-dimensional chaotic variables generated in step 4.1.2 to the solution search space according to the following formula, and the population Q0 after chaotic initialization is obtained:

[0125] x id = Lb + (Ub - Lb) x y id , d = 1, 2, 3, 4, 5

[0126] where Ub and Lb are the upper and lower limits of the dth dimension of the solution search space, y id is the dth dimension of the i th individual position generated in step (4.1.2), and x id is the dth coordinate value of the i th individual position in the solution search space.

[0127] (4.1.4) Determine whether each individual position in the initial population Q0 satisfies the processing constraints. If it does, it is retained. If it does not, the individual is removed and a new individual position is generated by applying a chaotic sequence. This process is repeated until the initial population Q0 contains N p individuals that satisfy the processing constraints.

[0128] (4.2) Calculate the fitness of each individual position in the population Q0, and check and mark each individual in the population whether it is dominated by other individuals.

[0129] (4.3) Update the individual positions in the population using the traditional Scarab Optimization Algorithm. Divide the individuals in the population Q t into four types according to their order in the population: ball-rolling Scarabs, brood Scarabs, small Scarabs, and thief Scarabs, and update them according to different strategies.

[0130] (4.3.1) The position update method for the first ball-rolling Scarab generated in step (4.3) is as follows:

[0131] (4.3.1.1) The ball-rolling Scarab will form the found fecal pile into a ball and push the fecal ball to the target position. The position update formula for the Scarab during ball rolling is as follows:

[0132] x i (t+1) = x i (t) + a x k x i (t-1) + b x Ax Ax = |x i (t) - x w |

[0133] where t represents the current iteration number; x i (t) represents the position of the ith dung beetle at the tth iteration; a is a natural coefficient representing the influence on the direction of pushing the ball, randomly selected as 1 or -1; k represents the position deflection coefficient, generally taking 0.1; b represents a random constant, generally taking 0.3; Ax is used to simulate the change of light intensity, and the higher the value of Ax, the weaker the light intensity; x w represents the global worst position.

[0134] (4.3.1.2) In addition, if the dung beetle encounters an obstacle during the process of pushing the ball, a deflection angle generated by a tangent function is used to obtain a new pushing route, and the position update formula at this time is as follows:

[0135] x i (t+1) = x i (t) + tan 0 x |x i (t) - x i (t-1)|

[0136] where 0 e [0, n], 0 represents a random number of the deflection angle of the dung beetle, and when 0 is 0, n / 2 or n respectively, the dung beetle does not update the position.

[0137] (4.3.1.3) Check the fitness of the position of each pushing dung beetle at present, and if the processing constraint is not met, the original individual position is retained, and if the processing constraint is met, the position is updated.

[0138] (4.3.2) The position update mode of the second kind of rearing dung beetle generated in step (4.3) is:

[0139] (4.3.2.1) In nature, in order to provide a safe environment for their offspring, it is essential for dung beetles to choose appropriate egg-laying sites. Inspired by the above behavior, a boundary selection strategy is proposed to simulate the area where the rearing dung beetle lays eggs, and the position update strategy is:

[0140] Lb * = max(X * x (1-R), Lb)

[0141] Ub * = min(X * x (1+R), Ub)

[0142] where X * represents the current local optimal position, Lb * and Ub * represent the lower and upper bounds of the oviposition region, respectively, R = 1 - t / T, t represents the current iteration number, T represents the maximum iteration number, Lb and Ub represent the lower and upper bounds of the optimization problem, respectively.

[0143] (4.3.2.2) Once the oviposition region is determined, the breeding scarab will choose the breeding ball in this region to lay eggs. In addition, it can be clearly seen from the position updating strategy that the boundary range of the oviposition region is dynamically changing, which is mainly determined by the value of R. Therefore, the position of the breeding scarab is also dynamic during the iteration process, which is represented as:

[0144] B i (t+1) = X * + b1 x (B i (t) - Lb * ) + b2 x (B i (t) - Ub * )

[0145] where B i (t) is the position information of the i-th breeding scarab at the t-th iteration, b1 and b2 represent two independent random vectors with a size of 1 x D, and D represents the dimension of the optimization problem. It is worth noting that the position of the breeding scarab needs to be strictly limited within a certain range, i.e. the oviposition region in step (4.3.2.1).

[0146] (4.3.2.3) Check the fitness of the current position of each breeding scarab, and if the processing constraint is not met, keep the original position, if the processing constraint is met, update the position.

[0147] (4.3.3) The position updating method of the third kind of small scarab generated in step (4.3) is:

[0148] (4.3.3.1) Some scarabs that have grown into adults crawl out of the ground to find food, which are called small scarabs. In addition, it is necessary to establish the best foraging area to guide the scarabs to forage, which simulates the foraging process of these scarabs in nature. Specifically, the boundary of the best foraging area is defined as follows:

[0149] Lb b = max(X b x (1 - R), Lb)

[0150] Ub b = min(X b x (1 + R), Ub)

[0151] where Xb represents the global best position, Lb b and Ub b respectively represent the lower and upper limits of the best foraging area.

[0152] (4.3.3.2) Therefore, the position of the small dung beetle is updated as follows:

[0153] x i (t+1) = x i (t) + C1x i (t) - Lb b ) + C2x i (t) - Ub b )

[0154] where x i (t) represents the position information of the i-th small dung beetle at the t-th iteration, C1 represents a random number subject to a normal distribution, and C2 represents a random vector belonging to (0, 1).

[0155] (4.3.3.3) Check the fitness of the current position of each small dung beetle. If the processing constraint is not met, the original position is retained. If the processing constraint is met, the position is updated.

[0156] (4.3.4) The position update method of the fourth kind of small thief dung beetle generated in step (4.3) is:

[0157] (4.3.4.1) In addition, there are some free-loading dung beetles in nature, which are called small thief dung beetles, and they steal dung balls from other dung beetles. Since X b is the best food source, it can be assumed that the vicinity of X b is the best place to compete for food, so during the iteration process, the position information of the thief dung beetle is updated as follows:

[0158] x i (t+1) = X b +Sxg(|x i (t) - X * | + |x i (t) - X b |)

[0159] where x i (t) represents the position information of the i-th thief dung beetle at the t-th iteration, g is a random vector of size 1xD subject to a normal distribution, and S is generally taken as 0.5.

[0160] (4.3.4.2) Check the fitness of the current position of each small dung beetle. If the processing constraint is not met, the original individual position is retained. If the processing constraint is met, the individual position is updated.

[0161] (4.4) After completing the above position update, the iteration count is incremented by one, i.e., t+1, at which point a new population Q is obtained. t First, it is necessary to determine whether the solution set corresponding to the position of each individual in the new population exceeds the limit of the cutting parameters. If it is higher than the limit, the upper limit of the cutting parameters is taken as the upper limit of the solution. If it is lower than the lower limit of the cutting parameters, the lower limit of the cutting parameters is taken as the lower limit of the solution. Then, the fitness of the positions of all individuals in the new population is calculated, and this is used to determine whether each individual is dominated by other individuals. Finally, the dominant individuals and dominated individuals of each individual are recorded.

[0162] (4.5) Merging populations Q t And Q t-1 And the individual positions of all individuals in the existing Pareto solution set, compare all non-dominated solutions recorded in steps (4.2) and (4.4), find all non-dominated solutions after merging the population, then sort all non-dominated solutions according to crowding distance, take the top N non-dominated solutions and add them to the Pareto solution set, and regard these solutions as the optimal Pareto solution set.

[0163] (4.6) Check if the maximum number of iterations T has been reached. If not, repeat steps (4.3) to (4.6) until the maximum number of iterations is reached. The Pareto solution set at this time is the optimal Pareto solution set. You can select a suitable combination of cutting parameters from it according to the actual application.

[0164] It should be noted that the embodiments described above are merely preferred embodiments of the present invention. For those skilled in the art, various modifications, improvements, and equivalent substitutions can be made to the present invention without departing from its principles, and such modifications, improvements, and equivalent substitutions are also considered to fall within the protection scope of the claims of the present invention.

Claims

1. A method for optimization of process parameters for fiber reinforced polymer composites considering the effect of tool wear throughout the milling cycle on machinability, characterized in that, The method comprises the following steps: (1) constructing a multiple regression fitting model of cutting force, machining quality, and layer damage about cutting progress according to experimental data of cutting machining; (2) determining constraint conditions of the fitting model according to performance of the numerical control machine tool and surface machining quality requirements of the machined part; (3) constructing a multi-objective cutting parameter optimization model in combination with the fitting model in step (1) and the constraint conditions in step (2); (4) solving the cutting parameter optimization model by using the improved scarab optimization algorithm; The step (4) specifically comprises the following sub-steps: (4.1) Set the basic parameters of the algorithm, including the population size N p , the number of solutions in the final output pareto solution set N, the proportion of the four-part population to all populations, the current iteration number t and the maximum iteration number T, while applying a chaotic sequence to generate the initial population Q t , t = 0, as follows: ， ，r = 4 ， d =1，2，3，4，5 wherein is the position of the n th individual, and are the upper and lower bounds, respectively, of the d th dimension of the solution search space, is the i th dimension of the position of the d th individual, is the i th coordinate value of the position of the d th individual in the solution search space. (4.2) calculating the corresponding fitness of the position of each individual in the population Q0, checking and marking whether each individual in the population is dominated by other individuals; (4.3) Apply the traditional dung beetle optimization algorithm to update the position of individuals in the population, and divide the individuals in the population Q t into the following four types according to the proportion in order: ball pushing dung beetle, breeding dung beetle, small dung beetle, and small thief dung beetle, and perform different strategies for updating the four types of dung beetles respectively; (4.3.1) the position updating mode of the push ball scarab is as follows: wherein, t represents the current iteration number; represents the position of the i M. ruficollis at the position of the t M. ruficollis at the position of the α randomly chosen 1 or -1; k taken 0.1; b taken 0.3; for simulating the change of light intensity; represents the global worst position; After encountering an obstacle, the position updating mode is as follows: wherein, ∈ [0, π], when When φ is 0, π / 2 or π, respectively, the dung beetle does not update the position; (4.3.2) the position updating mode of the breeding scarab is as follows: in, This indicates the current local optimal position. and These represent the lower and upper boundaries of the spawning area, respectively. T represents the maximum number of iterations. and Let these represent the lower bound and upper bound of the optimization problem, respectively. It is the first t During the nth iteration i Location information of the brooding dung beetles. b 1 and b 2 represents two independent random vectors of size 1×D, where D is the dimension of the optimization problem; (4.3.3) the position updating mode of the small scarab is as follows: wherein represents the globally optimal position, and respectively represent the lower and upper bounds of the optimal foraging area, represents the first i position information of the small harvester termites at the first t iteration, represents a random number subject to a normal distribution, represents a random vector belonging to (0, 1); (4.3.4) the position updating mode of the stealing scarab is as follows: wherein represents the i stealing only the position information of the t iterative, g is a random vector of size 1 x D subject to a normal distribution, S takes 0.5; (4.4) judging whether the solution set corresponding to the position of each individual in the new population exceeds the limit of the cutting parameter and taking corresponding measures, then calculating the fitness of the position of all individuals in the new population and judging whether each individual is dominated by other individuals, and recording the dominated individuals and dominated individuals of each individual; (4.5) merging population Q t and Q t-1 And the individual position of all individuals in the existing pareto solution set, compare all non-dominated solutions recorded in step (4.2) and step (4.4), find all non-dominated solutions after merging the population, then sort all non-dominated solutions according to the crowding distance, take the first N non-dominated solutions and put them into the pareto solution set, and these solutions are regarded as the optimal pareto solution set; (4.6) checking whether the maximum iteration number T is reached, if not, repeating steps (4.3) to (4.6) until the maximum iteration number is reached, at this time, the pareto solution set is the optimal pareto solution set.

2. The method for optimization of process parameters of fiber reinforced polymer composites considering the effect of tool wear on machinability throughout the milling cycle according to claim 1, characterized in that, The step (1) specifically comprises the following sub-steps: (1.1) Based on the axial depth of cut of the CNC machine tool Radial depth of cut Feed per tooth Spindle speed n And consider the changes in the entire milling cycle as well. L The influence of progressive tool wear on the surface roughness of the machined surface is investigated, and a fitting model for the surface roughness of the machined surface during the milling process is established. L For feed distance: wherein , a 1 、a 2 、a 3 、a 4 、a 5is a constant, is the surface roughness of the machined surface; (1.2) Based on the axial cutting depth of the CNC machine tool Radial depth of cut Feed per tooth Spindle speed n And consider the changes in the entire milling cycle as well. L The influence of progressive tool wear on the milling delamination damage factor is investigated, and a fitting model for the delamination damage factor during the milling process is established. wherein , b 1 、b 2 、b 3 、b 4 、b 5is a constant, D is a milling layer damage factor; (1.3) Based on the axial depth of cut of the CNC machine tool Radial depth of cut Feed per tooth Spindle speed n And consider the changes in the entire milling cycle as well. L The influence of progressive tool wear on milling force is investigated, and a fitting model of milling force during the milling process is established: wherein , d 1 、d 2 、d 3 、d 4 、d 5 is a constant, F is the milling force.

3. The method of optimization of process parameters for fibre reinforced polymer composites considering the effect of tool wear on machinability throughout the milling cycle as claimed in claim 1, wherein, The step (2) specifically comprises the following sub-steps: (2.1) determining the constraint of the spindle speed of the numerical control machine tool: wherein and are the lowest and highest rotational speeds allowed in the machine tool spindle or engineering. (2.2) determining the constraint of the feed per tooth of the numerical control machine tool: wherein , respectively indicate the minimum and maximum feed per tooth allowed in the machine or in the engineering; (2.3) determining the constraint of the axial cutting depth of the numerical control machine tool: wherein and respectively represent the minimum and maximum axial cut allowed in the machine or in the engineering. (2.4) determining the constraint of the radial cutting depth of the numerical control machine tool: wherein and respectively represent the minimum and maximum radial cut allowed in the machine or in the engineering. (2.5) determining the constraint of the surface roughness of the machined workpiece of the numerical control machine tool: In the formula an upper limit of the surface roughness required for the workpiece; (2.6) determining the constraint of the FRP milling layer damage factor: In the formula represents the upper limit of the FRP milling delamination damage factor (2.7) determining the constraint of the milling force: In the formula represents the maximum cutting force allowed by the machine tool feed mechanism.

4. The method for optimization of process parameters of fiber reinforced polymer composites considering the effect of tool wear on machinability throughout the milling cycle according to claim 1, characterized in that, The step (3) specifically comprises constructing the following parameter optimization model: S.t ; wherein is a machine tool cutting process efficiency objective function, z is the number of teeth of the tool; is a tool life objective function characterized by the amount of material removed.

Images